A Practical Method for Enantioselective Synthesis of All-Carbon QuaternaryStereogenic Centers through NHC-Cu-Catalyzed Conjugate Additions of Alkyl-

and Arylzinc Reagents to â-Substituted Cyclic Enones

Kang-sang Lee, M. Kevin Brown, Alexander W. Hird, and Amir H. Hoveyda*

Department of Chemistry, Merkert Chemistry Center, Boston College, Chestnut Hill, Massachusetts 02467

Received April 5, 2006; E-mail: amir.hoveyda@bc.edu

Catalytic asymmetric conjugate addition (ACA) of carbon-basednucleophiles to unsaturated carbonyls provides a direct route forsynthesis of enantiomerically enriched organic molecules.1 Progresshas been achieved in connection with catalytic ACA of alkylmetalsto different classes of unsaturated carbonyl compounds. Suchadvances include the development of amino acid-based chiralphosphines in these laboratories for Cu-catalyzed ACA of alkylzincreagents to cyclic2 and acyclic enones,3 nitroalkenes,4 and N-acyloxazolidinones.5 A related approach involves Cu- and Rh-catalyzed hydride conjugate additions toâ-substituted enones.1,6

The significant majority of the above investigations involvecatalytic enantioselective synthesis of tertiary C-C bonds. The morechallenging problem of catalytic ACAs that afford all-carbonquaternary stereogenic centers,7 a task that cannot be addressed byenantioselective hydride additions, has been the subject of only asmall cluster of disclosures. Alexakis reported in 2005 that chiralphosphoramidite‚Cu complexes can be utilized to catalyze ACAof Me3Al and Et3Al to â-alkyl-substituted cyclohexenones.8 Cata-lytic ACA of the less reactive but more functional group-tolerantdialkylzinc reagents have been limited to highly electrophilic(activated) substrates. We showed in 2005 that chiral amino acid-based ligands may be employed for Cu-catalyzed ACA of acyclictrisubstituted nitroalkenes9 and cyclic tetrasubstituted alkylideneâ-ketoesters.10 Carretero has used Rh-catalyzed ACA of alkenyl-boronic acids to unsaturated pyridyl sulfones (chiraphos as lig-and).11 Most recently, Fillion has outlined a method for Cu-cata-lyzed ACA of dialkylzinc reagents to acyclic aryl-substitutedalkylidene â-ketoesters that are derived from Meldrum’s acid(phosphoramidites again used as ligands).12 Our initial focus onreactions of the more activated systems (nitroalkenes and unsatur-atedâ-ketoesters) stemmed from the fact that peptide‚Cu complexesare ineffective in catalyzing ACAs of alkylzincs to simpleâ-substituted enones (<50% conv and<20% ee). To address theabove reactivity and selectivity problems, we turned our attentionto chiral bidentateN-heterocyclic carbenes (Scheme 1), entitiesdeveloped in these laboratories for Ru-catalyzed enantioselec-tive olefin metathesis13 and later applied to Cu-catalyzed allylicalkylations.14

Herein we report the first examples of Cu-catalyzed ACAs ofalkyl- and arylzinc reagents to simple unactivatedâ-substitutedcyclic enones. Transformations are promoted with of 2.5-15 mol% of a readily available chiral NHC-based Cu complex,15 and affordproducts that bear all-carbon quaternary stereogenic centers in67-98% yield and up to 97% ee. Catalytic reactions can be carriedout on a benchtop, with undistilled solvent and commerciallyavailable (not purified) Cu salts and in reasonable scale. Tentativemechanistic models, accounting for the observed levels and trendsin enantioselectivity, are provided.

Preliminary investigations focused on chiral NHC complexes1-4 (Scheme 1); conversion of cyclohexenone5 to ketone6 served

as the representative process (Table 1). These studies indicated that,although inefficient (32% conv, 48 h,-30 °C), Cu(II) complex1promotes the ACA with appreciable enantioselectivity (72% ee;entry 1, Table 1). The “second-generation” NHC‚Cu complex2proved to be more effective (53% conv, 48 h), delivering6 in 59%ee. Next, we investigated NHC‚Ag(I) complex4, which is precursorto 2,14 and the ability of the derived in situ generated NHC‚Cucomplexes to promote ACA. As illustrated in entries 3-7 (Table1), in all cases, including air-stable CuOAc and CuTC salts (entries5-6), reactions proceed readily (93-98% conv in 6 h), and thedesired product is isolated with high asymmetric induction. Thehighest enantioselectivity is obtained with (CuOTf)2‚C6H6, used inconjunction with4, affording6 efficiently (94% conv in 6 h; 92%

Scheme 1

Table 1. Initial Screening Studiesa

entry catalyst mol (%) time (h) conv (%)b ee (%)c

1 (S,S)-1 2.5 48 32 722 (S,S)-2 2.5 48 53 593 (S,S)-4; CuCl2‚2H2O 2.5; 5 48 56 484 (S,S)-4; Cu(OTf)2 2.5; 5 48 45 785 (S,S)-4; CuOAc 2.5; 5 6 98 896 (S,S)-4; CuTC 2.5; 5 6 93 917 (S,S)-4; (CuOTf)2‚C6H6 2.5; 2.5 6 94 938 (S,S)-3; (CuOTf)2‚C6H6 2.5; 2.5 48 85 96

a Reactions performed under N2 atm. b Determined by1H NMR analysis.c Determined by chiral GLC analysis (see the SI for details).

Published on Web 05/13/2006

7182 9 J. AM. CHEM. SOC. 2006 , 128, 7182-7184 10.1021/ja062061o CCC: $33.50 © 2006 American Chemical Society

yield) and in 93% ee. Use of NHC-Ag(I) complex3 and (CuOTf)2‚C6H6 also leads to improvement in reactivity and selectivity (entry8 vs with NHC-Cu(II) complex1 in entry 1); the combination inentry 7, however, delivers a more facile ACA. The data summarizedin Table 1 indicate that, when Cu(I) salts are used to generate thechiral complex (entries 5-6), higher catalyst efficiency is achieved(vs Cu(II) salts;>98% conv for entries 5-6 vs 45-55% conv forentries 3-4). This difference in reactivity may arise from the slowrate of Cu(II)f Cu(I) reduction under the reaction conditions, orcould be due to low solubility of Cu(II) salts in Et2O.

The combination of NHC‚Ag(I) complex4 and (CuOTf)2‚C6H6

can be utilized to promote catalytic ACAs of commercially available(not purified) dialkylzinc reagents toâ-alkyl- andâ-aryl-substitutedcyclic enones in 54-95% ee (Table 2).

Several points regarding the data in Table 2 are noteworthy: (1)Transformations typically proceed to>80% conv after 6-48 h(-30 °C); comparison between percent conversion and isolatedyield demonstrates that there is minimal formation of byproducts(e.g., 1,2-addition). (2) Rates of reactions and levels of enantio-selectivity are sensitive to steric factors. There is<5% conversionwhen (i-Pr)2Zn is used. Catalytic ACA of Me-substituted enone(6) in entry 1 (Table 2) requires 2.5 mol %4 (94% conv), affordingthe desired product in 93% ee; 5 mol % catalyst is needed forn-alkenyl-substituted enone in entry 3 to proceed to 93% conversion,delivering8 in 84% ee. Consistent with this trend, enantioslectivesynthesis ofâ-phenyl-substituted10 (90% ee, 85% conv) must be

performed with 10 mol %4. (3) Catalytic ACAs of cycloheptenones(entries 6-8, Table 2) deliver cyclic enones11-13 in 76-85% eeand 67-83% isolated yield. However, these transformations areslower than their six-membered ring analogues. (4) Reactions ofeight-membered rings are slower and less selective thanâ-sub-stituted cyclohexenones and cycloheptenones; the example pro-vided in entry 9 (Table 2) is illustrative. The relatively unreactiveMe2Zn cannot be used (<5% conv, 48 h), and there is<5%conversion with cyclopentenones (after 48 h).

The Cu-catalyzed protocol can be used to promote ACA ofdiarylzinc reagents; desired cyclic ketones are obtained in 88-95%isolated yield and 89-97% ee (Table 3). Transformations withPh2Zn proceed less readily than those involving dialkylzinc reagents,but with higher enantioselectivity (e.g., compare entries 1-3 ofTable 3 with entries 1-3 of Table 2). Cu-catalyzed ACA can beextended to an electron-rich diarylzinc reagent (entry 4 of Table3), albeit with lower enantioselectivity (90% ee vs 97% ee in entry1). Attempts to effect addition with the corresponding (p-CF3C6H4)2-Zn16 gave rise to<5% conversion (e.g., with enone5). As theexample in entry 5 of Table 3 indicates (18 formed in 96% ee),ACA of Ph2Zn to â-substituted cycloheptenones proceeds withenantioselectivity levels similar to those observed with cyclohex-enones. To the best of our knowledge, catalytic ACA transforma-tions of an arylmetal to afford an all-carbon quaternary stereogeniccenter have not been reported previously.17

Cu-catalyzed ACA of dialkylzinc reagents (Table 2) proceed inthe oppositesense compared to reactions with diarylzinc reagents(e.g., compare entry 5 of Table 2 and entry 2 of Table 3). Pre-liminary mechanistic models that account for the observed enan-tioselectivity, as well as the aforementioned reversal in the senseof asymmetric induction, are presented in Scheme 2.18 Alkyl-cuprateNHC complexes may undergo ACA throughI ; unfavorable stericinteraction involving the enone substituent (Me) and the NHC’sbiphenol moiety is therefore avoided. In contrast, with aryl-cupratecomplexes, the steric interaction between the enone’sâ-substituentand the aryl-Cu unit might be the dominant factor, giving rise topreference for complexII .

The optically enriched Zn-enolate intermediates can be convertedto versatile enolsilane or enoltriflate derivatives for further func-tionalization.19 Representative cases are depicted in Scheme 3.

Table 2. Cu-Catalyzed ACA of Dialkylzinc Reagents toâ-Substituted Cyclic Enonesa

a See Table 1 for conditions; 6 h for entry 1, 24-48 h otherwise.b By1H NMR analysis.c Isolated yields.d Determined by chiral GLC analysis(see the SI for details).e Reaction performed at-15 °C.

Table 3. Cu-Catalyzed ACA of Diarylzinc Reagents toâ-Substituted Cyclic Enonesa

a-d See Table 2; 48 h.e At -15 °C. f At -15 °C, in toluene, 72 h.g Time) 72 h.

C O M M U N I C A T I O N S

J. AM. CHEM. SOC. 9 VOL. 128, NO. 22, 2006 7183

Enolsilane19 is obtained with>98% regioselectivity and in>98%yield when the mixture from Cu-catalyzed ACA of Ph2Zn to 5 istreated with TMSOTf. Regioselective deprotonation of15 withLiTMP delivers enolsilane20 in 96% isolated yield and with 88:12 regioselectivity. Similarly,5 is converted to enoltriflate21(>98% yield), which can be subjected to Pd-catalyzed cross-coupling reactions.20 Conversion to cyclohexene22 in 87% isolatedyield serves as a case in point (Scheme 3).

The present class of transformations can be performed underoperationally simple conditions. As illustrated in eq 1,reactionscan be carried out with undistilled solVent and set up on benchtop(Schlenck-ware not needed).Catalytic ACA can be effected withcommercially available (CuOTf)2‚toluene; nonetheless, use offreshly prepared (CuOTf)2‚C6H6 leads to higher conversion.Although this study was focused on ACA with (CuOTf)2‚C6H6 (formaximum activity), as the examples in entries 5-6 of Table 1indicate, more user-friendly Cu salts are effective. When thecatalytic ACA in eq 1 is carried out with commercially availableand air stable CuOAc,6 is isolated in 86% yield and 90% ee. Cu-catalyzed additions can be performed on reasonable scale; thetransformation in eq 1, at 0.5-g scale ((CuOTf)2‚C6H6; -40 °C,8-12 h), delivers6 in 87% ee and in quantitative yield.

In summary, we have developed the first Cu-catalyzed ACA ofalkyl- and arylzinc reagents to unactivatedâ-substituted cyclicenones; the catalytic protocol is operationally straightforward anddelivers cyclic ketones that bear all-carbon quaternary stereogeniccenters in excellent yields and 54-97% ee. This is the firstapplication of this class of chiral bidentate NHC ligands to catalyticACA reactions.

Acknowledgment. Financial support was provided by the NIH(GM-47480) and the NSF (CHE-0213009). M.K.B. is a recipientof an ACS Graduate Fellowship in Organic Chemistry (2005-6,sponsored by Schering-Plough).

Supporting Information Available: Experimental procedures andspectral, analytical data for all reaction products. This material isavailable free of charge via the Internet at http://www.pubs.acs.org

References

(1) (a) Krause, N.; Hoffmann-Ro¨der, A. Synthesis2001, 171-196. (b)Alexakis, A.; Benhaim, C.Eur. J. Org. Chem.2002, 3221-3236. (c)Feringa, B. L.; Naasz, R.; Imbos, R.; Arnold, L. A. InModernOrganocopper Chemistry; Krause, N., Ed.; Wiley-VCH: Weinheim, 2002;pp 224-258.

(2) (a) Degrado, S. J.; Mizutani, H.; Hoveyda, A. H.J. Am. Chem. Soc.2001,123, 755-756. (c) Degrado, S. J.; Mizutani, H.; Hoveyda, A. H.J. Am.Chem. Soc.2002, 124, 13362-13363. (c) Brown, M. K.; Degrado, S. J.;Hoveyda, A. H.Angew. Chem., Int. Ed.2005, 44, 5306-5310. For anoverview, see: (d) Hoveyda, A. H.; Hird, A. W.; Kacprzynski, M. A.Chem. Commun.2004, 1779-1785.

(3) (a) Mizutani, H.; Degrado, S. J.; Hoveyda, A. H.J. Am. Chem. Soc.2002,124, 779-781. (b) Cesati, R. R., III; de Armas, J.; Hoveyda, A. H.J.Am. Chem. Soc.2004, 126, 96-101.

(4) (a) Luchaco-Cullis, C. A.; Hoveyda, A. H.J. Am. Chem. Soc.2002, 124,8192-8193. (b) Mampreian, D. M.; Hoveyda, A. H.Org. Lett.2004, 6,2829-2832.

(5) Hird, A. W.; Hoveyda, A. H.Angew. Chem., Int. Ed.2003, 42, 1276-1279.

(6) Hayashi, T.; Yamasaki, K.Chem. ReV. 2003, 103, 2829-2844.(7) (a) Denissova, I.; Barriault, L.Tetrahedron2003, 59, 10105-10146. (b)

Douglas, C. J.; Overman, L. E.Proc. Natl. Acad. Sci. U.S.A.2004, 101,5363-5367.

(8) d’Augustin, M.; Palais, L.; Alexakis, A.Angew. Chem., Int. Ed.2005,44, 1376-1378.

(9) Wu, J.; Mampreian, D. M.; Hoveyda, A. H.J. Am. Chem. Soc.2005,127, 4584-4585.

(10) Hird, A. W.; Hoveyda, A. H.J. Am. Chem. Soc.2005, 127, 14988-14989.

(11) Mauleon, P.; Carretero, J. C.Chem. Commun.2005, 4961-4963.(12) Fillion, E.; Wilsily, A. J. Am. Chem. Soc.2006, 128, 2829-2844.(13) (a) Van Veldhuizen, J. J.; Garber, S. B.; Kingsbury, J. S.; Hoveyda, A.

H. J. Am. Chem. Soc.2002, 124, 4954-4955. (b) Van Veldhuizen, J. J.;Gillingham, D. G.; Garber, S. B.; Kataoka, O.; Hoveyda, A. H.J. Am.Chem. Soc.2003, 125, 12502-12508. (c) Gillingham, D. G.; Kataoka,O.; Garber, S. B.; Hoveyda, A. H.J. Am. Chem. Soc.2004, 126, 12288-12290. (d) Van Veldhuizen, J. J.; Campbell, J. E.; Giudici, R. E.; Hoveyda,A. H. J. Am. Chem. Soc.2005, 127, 6877-6882.

(14) (a) Larsen, A. O.; Leu, W.; Nieto-Oberhuber, C.; Campbell, J. E.; Hoveyda,A. H. J. Am. Chem. Soc.2004, 126, 11130-11131. (b) Reference 13d.

(15) For catalytic ACAs of Et2Zn to disubstituted enones with NHC ligands,see: (a) Guillen, F.; Winn, C. L.; Alexakis, A.Tetrahedron: Asymmetry2001, 12, 2083-2086. (b) Pytkowicz, J.; Roland, S.; Mangeney, P.Tetrahedron: Asymmetry2001, 12, 2087-2089. (c) Alexakis, A.; Winn,C. L.; Guillen, F.; Pytkowicz, J.; Roland, S.; Mangeney, P.AdV. Synth.Catal. 2003, 345, 345-348. (d) Arnold, P. L.; Rodden, M.; Davis, K.M.; Scarisbrick, A. C.; Blake, A. J.; Wilson, C.Chem. Commun.2004,1612-1613. (e) Clavier, H.; Coutable, L.; Guillemin, J.-C.; Mauduit, M.Tetrahedron: Asymmetry2005, 16, 921-924.

(16) See the Supporting Information for experimental details.(17) For Cu-catalyzed ACAs of Ph2Zn to the more reactive disubstituted enones,

see: (a) Schinneri, M.; Seitz, M.; Kaiser, A.; Reiser, O.Org. Lett.2001,3, 4259-4262. (b) Pena, D.; Lopez, F.; Harutyunyan, S. R.; Minnaard,A. J.; Feringa, B. L.Chem. Commun.2004, 1836-1837. (c) Shi, M.;Wang, C.-J.; Zhang, W.Chem. Eur. J.2004, 10, 5507-5516.

(18) Although rigorous evidence is not yet available, due to consideration ofsteric factors, it is likely that monomeric NHC complex is the activecatalyst. The models proposed herein are preliminary and tentative; relatedhigh-level calculations are being carried out, the results of which will bereported in due course.

(19) For functionalization procedures involving enolsilane and enoltriflatesderived from catalytic ACA reactions, see: Knopff, O.; Alexakis, A.Org.Lett. 2002, 4, 3835-3837 and references therein.

(20) (a) Ritter, K.Synthesis1993, 8, 735-763. (b) Suarez, R. M.; Pena, D.;Minnaard, A. J.; Feringa, B. L.Org. Biomol. Chem.2005, 3, 729-731.

JA062061O

Scheme 2. Proposed Mechanistic Models

Scheme 3. Representative Functionalizations of ACA Products

C O M M U N I C A T I O N S

7184 J. AM. CHEM. SOC. 9 VOL. 128, NO. 22, 2006